fix: fee calculation formula

[ArielElp]

Description of the Changes

More adjustments to the fee formulate following some discrepancies from the actual calculation that happens in the blockifier.

PR Preview URL

Gas and transaction fees

Check List

• Changes have been done against main branch, and PR does not conflict
• PR title follows the convention: <docs/feat/fix/chore>(optional scope): <description>, e.g: fix: minor typos in code

---

This change is 

[github-actions]

Your preview build is ready! ✨ Check the following link in 1-2 minutes: https://starknet-io.github.io/starknet-docs/pr-1092/documentation/ .

[github-actions]

Your preview build is ready! ✨ Check the following link in 1-2 minutes: https://starknet-io.github.io/starknet-docs/pr-1092/documentation/ .

[github-actions]

Your preview build is ready! ✨ Check the following link in 1-2 minutes: https://starknet-io.github.io/starknet-docs/pr-1092/documentation/ .

[github-actions]

Your preview build is ready! ✨ Check the following link in 1-2 minutes: https://starknet-io.github.io/starknet-docs/pr-1092/documentation/ .

[github-actions]

Your preview build is ready! ✨ Check the following link in 1-2 minutes: https://starknet-io.github.io/starknet-docs/pr-1092/documentation/ .

[github-actions]

Your preview build is ready! ✨ Check the following link in 1-2 minutes: https://starknet-io.github.io/starknet-docs/pr-1092/documentation/ .

[stoobie]

I asked two folks to take a look at this, and one of them didn't understand what k is. The other person explained that k is a number that corresponds to one of the resources listed in Table 1, so k is a value 1-8, (or 0-7?). Which means that the table should have a column for resource identifier k.

[github-actions]

Your preview build is ready! ✨ Check the following link in 1-2 minutes: https://starknet-io.github.io/starknet-docs/pr-1092/documentation/ .

[github-actions]

Your preview build is ready! ✨ Check the following link in 1-2 minutes: https://starknet-io.github.io/starknet-docs/pr-1092/documentation/ .

[github-actions]

Your preview build is ready! ✨ Check the following link in 1-2 minutes: https://starknet-io.github.io/starknet-docs/pr-1092/documentation/ .

[github-actions]

Your preview build is ready! ✨ Check the following link in 1-2 minutes: https://starknet-io.github.io/starknet-docs/pr-1092/documentation/ .

[ArielElp]

Reviewed 1 of 1 files at r6.
Reviewable status: 0 of 1 files reviewed, 7 unresolved discussions (waiting on @stoobie)

---

components/Starknet/modules/architecture_and_concepts/pages/Network_Architecture/fee-mechanism.adoc line 31 at r1 (raw file):

Previously, stoobie (Steve Goodman) wrote…

* stem:[$m$] is the number of values updated, not counting multiple updates for the same key. Notice that stem:[$m\ge 2$] is always true, because the balance of the account sending the transaction and the sequencer's balance are always updated.

Added "always holds"

---

components/Starknet/modules/architecture_and_concepts/pages/Network_Architecture/fee-mechanism.adoc line 35 at r1 (raw file):

Previously, stoobie (Steve Goodman) wrote…

* stem:[$\text{calldata_cost}$] is 550 gas per 32-byte word. This cost is derived as follows: 512 gas per 32-byte word for calldata, ~100 gas for onchain hashing that happens for every word sent, and a 10% discount that applies because the sequencer does not incur additional costs for repeated updates to the same storage slot within a single block. So the gas cost is calculated as follows: stem:[0.9 \cot 612 = 550 gas], rounded down.

It's eventually 551 not 550. I'ts not rounded up, but rather we get it by doing 512-512*10/100 where both sides are integers, I don't want to get into it here.

---

components/Starknet/modules/architecture_and_concepts/pages/Network_Architecture/fee-mechanism.adoc line 197 at r1 (raw file):

Previously, stoobie (Steve Goodman) wrote…

Taking into account that zero bytes only cost 4 gas, the cost difference between a full 32-byte word, which is entirely non-zero, and a word with only six non-zero bytes is stem:[$32\cdot16-(6\cdot16+26\cdot4)=312$].

Done.

---

components/Starknet/modules/architecture_and_concepts/pages/Network_Architecture/fee-mechanism.adoc line 27 at r2 (raw file):

Previously, stoobie (Steve Goodman) wrote…

I asked two folks to take a look at this, and one of them didn't understand what k is. The other person explained that k is a number that corresponds to one of the resources listed in Table 1, so k is a value 1-8, (or 0-7?). Which means that the table should have a column for resource identifier k.

I don't want to explicitly refer to it, it is indexing the entries of v and w, so if you go to read about v and w you understand what the entries are (it is also mentioned in the first point)

---

components/Starknet/modules/architecture_and_concepts/pages/Network_Architecture/fee-mechanism.adoc line 188 at r2 (raw file):

Previously, stoobie (Steve Goodman) wrote…

* stem:[$m$] is the number of values updated, not counting multiple updates for the same key. Notice that stem:[$m\ge 2$] is always true,  because the balance of the account sending the transaction and the sequencer's balance are always updated. The sender's balance is treated separately, as described below, and the sequencer's balance is ignored since it is touched by all the transactions within a block.

Added "always holds"

---

components/Starknet/modules/architecture_and_concepts/pages/Network_Architecture/fee-mechanism.adoc line 189 at r2 (raw file):

Previously, stoobie (Steve Goodman) wrote…

* stem:[\text{contract_update_discount}] is 312 gas, which is discounted for every updated contract. This discount is a result of the fact that out of the stem:[$2n$] words caused by updating contracts, stem:[$n$] words are short, including at most 6 non-zero bytes:

Done.

---

components/Starknet/modules/architecture_and_concepts/pages/Network_Architecture/fee-mechanism.adoc line 198 at r2 (raw file):

Previously, stoobie (Steve Goodman) wrote…

* stem:[$310$] is the cost of updating the sender's balance, and is derived by assuming the balance requires at most 12 non-zero bytes, which is enough for 1.2B ETH or STRK, resulting in the following discounted price: stem:[$0.9 \cdot 612-(512-(16 \cdot12+20 \cdot 4))=310$], rounded down.

Done.

[github-actions]

Your preview build is ready! ✨ Check the following link in 1-2 minutes: https://starknet-io.github.io/starknet-docs/pr-1092/documentation/ .

[github-actions]

Your preview build is ready! ✨ Check the following link in 1-2 minutes: https://starknet-io.github.io/starknet-docs/pr-1092/documentation/ .

[github-actions]

Your preview build is ready! ✨ Check the following link in 1-2 minutes: https://starknet-io.github.io/starknet-docs/pr-1092/documentation/ .

[github-actions]

Your preview build is ready! ✨ Check the following link in 1-2 minutes: https://starknet-io.github.io/starknet-docs/pr-1092/documentation/ .

[github-actions]

Your preview build is ready! ✨ Check the following link in 1-2 minutes: https://starknet-io.github.io/starknet-docs/pr-1092/documentation/ .

[github-actions]

Your preview build is ready! ✨ Check the following link in 1-2 minutes: https://starknet-io.github.io/starknet-docs/pr-1092/documentation/ .

[github-actions]

Your preview build is ready! ✨ Check the following link in 1-2 minutes: https://starknet-io.github.io/starknet-docs/pr-1092/documentation/ .

[github-actions]

Your preview build is ready! ✨ Check the following link in 1-2 minutes: https://starknet-io.github.io/starknet-docs/pr-1092/documentation/ .

[github-actions]

Your preview build is ready! ✨ Check the following link in 1-2 minutes: https://starknet-io.github.io/starknet-docs/pr-1092/documentation/ .

[github-actions]

Your preview build is ready! ✨ Check the following link in 1-2 minutes: https://starknet-io.github.io/starknet-docs/pr-1092/documentation/ .

[stoobie]

Reviewable status: 0 of 1 files reviewed, 14 unresolved discussions (waiting on @ArielElp)

---

components/Starknet/modules/architecture_and_concepts/pages/Network_Architecture/fee-mechanism.adoc line 31 at r1 (raw file):

Previously, ArielElp wrote…

Added "always holds"

I changed the sentence to

* stem:[$m$] is the number of values updated, not counting multiple updates for the same key. Notice that stem:[$m\ge 1$] is always true, because the sequencer's balance is always updated, which does not incur any fee.

---

components/Starknet/modules/architecture_and_concepts/pages/Network_Architecture/fee-mechanism.adoc line 35 at r1 (raw file):

Previously, ArielElp wrote…

It's eventually 551 not 550. I'ts not rounded up, but rather we get it by doing 512-512*10/100 where both sides are integers, I don't want to get into it here.

So is this what you want to say (with my edits):

* stem:[$\text{da_calldata_cost}$] is 551 gas per 32-byte word. This cost is derived as follows: 
+
** 512 gas per 32-byte word for calldata.
** ~100 gas for onchain hashing that happens for every sent word.
** a 10% discount, because the sequencer does not incur additional costs for repeated updates to the same storage slot within a single block.

---

components/Starknet/modules/architecture_and_concepts/pages/Network_Architecture/fee-mechanism.adoc line 188 at r2 (raw file):

Previously, ArielElp wrote…

Added "always holds"

Please approve my edits.

---

components/Starknet/modules/architecture_and_concepts/pages/Network_Architecture/fee-mechanism.adoc line 36 at r9 (raw file):

+
For more information see xref:#calculation_of_computation_costs[Calculation of computation costs].
* stem:[$n$] is xref:#storage_updates[the number of unique contracts updated]. Note that this also takes into account contract deployment, even if the storage of the newly deployed contract was untouched, i.e. stem:[$n\ge\ell$].

Does "the number of unique contracts updated" include "the number of contracts whose class was changed"? It's not clear, but I think that's what you mean. Assuming I'm correct, then the following text should be correct:

• stem:[$n$] is xref:#storage_updates[the number of unique contracts updated], which includes changes to, or new deployments of, classes of existing contract instances, even if the storage of the newly deployed contract is untouched. In other words, stem:[$n\ge\ell$].

Code quote:

* stem:[$n$] is xref:#storage_updates[the number of unique contracts updated]. Note that this also takes into account contract deployment, even if the storage of the newly deployed contract was untouched, i.e. stem:[$n\ge\ell$].

---

components/Starknet/modules/architecture_and_concepts/pages/Network_Architecture/fee-mechanism.adoc line 39 at r9 (raw file):

* stem:[$m$] is the number of values updated, not counting multiple updates for the same key. Note that stem:[$m\ge 1$] always holds since the sequencer balance is always updated (this is not charged for).
* stem:[$t$] is the number of L2->L1 messages sent, where the corresponding payload sizes are denoted by stem:[$q_1,...,q_t$].
* stem:[$\ell$] is the number of contracts whose class was changed. This can happen on deployment or on applying the `replace_class` syscall.

s whose class was changed, which can happen on deployment of a new version of the class, or on applying the replace_class syscall.

Code quote:

s whose class was changed. This can happen on deployment or on applying the `replace_class` syscall.

---

components/Starknet/modules/architecture_and_concepts/pages/Network_Architecture/fee-mechanism.adoc line 248 at r14 (raw file):

* class hash

The first two elements are counted in the number of unique modified contracts, denoted by stem:[$n$] throughout this page, such as with stem:[$n\ge\ell$]. So the only additional comes from publishing the class hash,

 So the only additional cost comes from publishing the class hash,

Code quote:

 So the only additional comes from publishing the class hash,

[github-actions]

Your preview build is ready! ✨ Check the following link in 1-2 minutes: https://starknet-io.github.io/starknet-docs/pr-1092/documentation/ .

[ArielElp]

Reviewed 1 of 1 files at r13.
Reviewable status: 0 of 1 files reviewed, 14 unresolved discussions (waiting on @stoobie)

---

components/Starknet/modules/architecture_and_concepts/pages/Network_Architecture/fee-mechanism.adoc line 31 at r1 (raw file):

Previously, stoobie (Steve Goodman) wrote…

I changed the sentence to

* stem:[$m$] is the number of values updated, not counting multiple updates for the same key. Notice that stem:[$m\ge 1$] is always true, because the sequencer's balance is always updated, which does not incur any fee.

slightly changed

---

components/Starknet/modules/architecture_and_concepts/pages/Network_Architecture/fee-mechanism.adoc line 35 at r1 (raw file):

Previously, stoobie (Steve Goodman) wrote…

So is this what you want to say (with my edits):

* stem:[$\text{da_calldata_cost}$] is 551 gas per 32-byte word. This cost is derived as follows: 
+
** 512 gas per 32-byte word for calldata.
** ~100 gas for onchain hashing that happens for every sent word.
** a 10% discount, because the sequencer does not incur additional costs for repeated updates to the same storage slot within a single block.

Done.

---

components/Starknet/modules/architecture_and_concepts/pages/Network_Architecture/fee-mechanism.adoc line 188 at r2 (raw file):

Previously, stoobie (Steve Goodman) wrote…

Please approve my edits.

Done.

---

components/Starknet/modules/architecture_and_concepts/pages/Network_Architecture/fee-mechanism.adoc line 36 at r9 (raw file):

Previously, stoobie (Steve Goodman) wrote…

Does "the number of unique contracts updated" include "the number of contracts whose class was changed"? It's not clear, but I think that's what you mean. Assuming I'm correct, then the following text should be correct:

• stem:[$n$] is xref:#storage_updates[the number of unique contracts updated], which includes changes to, or new deployments of, classes of existing contract instances, even if the storage of the newly deployed contract is untouched. In other words, stem:[$n\ge\ell$].

Changed it slightly

---

components/Starknet/modules/architecture_and_concepts/pages/Network_Architecture/fee-mechanism.adoc line 37 at r9 (raw file):

Previously, stoobie (Steve Goodman) wrote…

* stem:[$m$] is the number of values updated, not counting multiple updates for the same key. Notice that stem:[$m\ge 1$] is always true, because the sequencer's balance is always updated, which does not incur any fee.

Done.

---

components/Starknet/modules/architecture_and_concepts/pages/Network_Architecture/fee-mechanism.adoc line 39 at r9 (raw file):

Previously, stoobie (Steve Goodman) wrote…

* stem:[$\ell$] is the number of contracts whose class was changed, which can happen on deployment of a new version of the class, or on applying the `replace_class` syscall.

This is not true, changed it back to either contract deployment of class replacement

---

components/Starknet/modules/architecture_and_concepts/pages/Network_Architecture/fee-mechanism.adoc line 39 at r9 (raw file):

Previously, stoobie (Steve Goodman) wrote…

s whose class was changed, which can happen on deployment of a new version of the class, or on applying the replace_class syscall.

Done.

---

components/Starknet/modules/architecture_and_concepts/pages/Network_Architecture/fee-mechanism.adoc line 41 at r9 (raw file):

Previously, stoobie (Steve Goodman) wrote…

* stem:[$\text{da_calldata_cost}$] is 551 gas per 32-byte word. This cost is derived as follows: 
+
** 512 gas per 32-byte word for calldata.
** ~100 gas for onchain hashing that happens for every sent word.
** a 10% discount, because the sequencer does not incur additional costs for repeated updates to the same storage slot within a single block.

Done.

---

components/Starknet/modules/architecture_and_concepts/pages/Network_Architecture/fee-mechanism.adoc line 195 at r9 (raw file):

Previously, stoobie (Steve Goodman) wrote…

* stem:[$m$] is the number of values updated, not counting multiple updates for the same key. Notice that stem:[$m\ge 1$] is always true, because the sequencer's balance is always updated, which does not incur any fee.

Changed to the same thing I did above

---

components/Starknet/modules/architecture_and_concepts/pages/Network_Architecture/fee-mechanism.adoc line 224 at r9 (raw file):

Previously, stoobie (Steve Goodman) wrote…

Consequently, the gas cost associated with a single L2→L1 message is:

Done.

---

components/Starknet/modules/architecture_and_concepts/pages/Network_Architecture/fee-mechanism.adoc line 231 at r9 (raw file):

Previously, stoobie (Steve Goodman) wrote…

Where stem:[$\text{da_calldata_cost}$] is 551 gas, and stem:[$\text{message_calldata_cost}$] is 512 gas.

Done.

---

components/Starknet/modules/architecture_and_concepts/pages/Network_Architecture/fee-mechanism.adoc line 36 at r10 (raw file):

Previously, stoobie (Steve Goodman) wrote…

* stem:[$n$] is xref:#storage_updates[the number of unique contracts updated], which includes changes to, or new deployments of, classes of existing contract instances, even if the storage of the newly deployed contract is untouched. In other words, stem:[$n\ge\ell$].

Done.

---

components/Starknet/modules/architecture_and_concepts/pages/Network_Architecture/fee-mechanism.adoc line 198 at r10 (raw file):

Previously, stoobie (Steve Goodman) wrote…

* stem:[$n$] is xref:#storage_updates[the number of unique contracts updated], which includes changes to, or new deployments of, classes of existing contract instances, even if the storage of the newly deployed contract is untouched. In other words, stem:[$n\ge\ell$].

Done.

---

components/Starknet/modules/architecture_and_concepts/pages/Network_Architecture/fee-mechanism.adoc line 249 at r11 (raw file):

Previously, stoobie (Steve Goodman) wrote…

which adds 551 gas. For more information, see stem:[$\text{da_calldata_cost}$] and stem:[$\text{message_calldata_cost}$] in xref:#l_2-l_1_messages[Onchain data: L2->L1 messages].

This is unrelated to messages, changed the link to the final formula

[github-actions]

Your preview build is ready! ✨ Check the following link in 1-2 minutes: https://starknet-io.github.io/starknet-docs/pr-1092/documentation/ .

[github-actions]

Your preview build is ready! ✨ Check the following link in 1-2 minutes: https://starknet-io.github.io/starknet-docs/pr-1092/documentation/ .

[github-actions]

Your preview build is ready! ✨ Check the following link in 1-2 minutes: https://starknet-io.github.io/starknet-docs/pr-1092/documentation/ .

[stoobie]

Reviewable status: 0 of 1 files reviewed, all discussions resolved (waiting on @ArielElp)